We are not worthy of Artificial Intelligence, but that shouldn’t matter

Reflecting, in part, on the threatening development of nuclear technology during the 1930s, French biologist Jean Rostand lamented that “science has made us gods before we are worthy of being human.”

It was a prescient observation, and one that summarizes well the innumerable articles published in recent months warning about the apocalyptic dangers inherent in our development of artificial intelligence, or AI.

Sundar Pichai, CEO of Google parent company Alphabet, has even opined that AI could be more significant to humanity than “fire or electricity.” And while it stands to reason that AI will, undoubtedly, be used for nefarious purposes, we need to calm down. Sure, it may do some bad, but AI will likely do far more for humanity that is good, as was the case with our application of fire and electricity.

It is impossible to apply an evolving technology “perfectly” without trial and error. Instead, as humans, we learn as we go. Prior to the introduction of electricity, it would have been impossible to imagine all the various ways electricity might be used to commit evil deeds.

For example, it is unlikely that Thomas Edison foresaw phishing scams that are entirely reliant upon electrical equipment. But just because a technological advance can be used for ill-gotten gains in unimaginable ways, that does not mean we should halt its development entirely until humanity perfects itself.

Should we have halted the use of fire out of fear that a percentage of our ancestors were arsonists? Probably not. Because if human ethical perfection is the prerequisite to engaging in science, we might as well forgo developing new technology altogether.

Others in the AI community are less-concerned with human malpractice, and are instead focused on the computers making bad actions. "In these scenarios, AI isn’t necessarily sentient. Instead, it becomes fixated on a goal — even a mundane one, like making paper clips — and triggers human extinction to optimize its task."

However far-fetched it may sound, we can work to overcome and prevent these scenarios. Plus, we have a lot of experience with AI already.

Artificial intelligence, by definition, is a type of technology that evolves. It is “a machine’s ability to perform the cognitive functions we associate with human minds, such as perceiving, reasoning, learning, interacting with an environment, problem solving, and even exercising creativity,” per McKinsey.  

But there’s a bit of a misperception that AI is wholly new and unknown, (in part because of recent headlines that suggest AI will be more important to humans than fire and electricity). AI is not a new technology that spontaneously appeared in 2023. Our ability to create artificial intelligence has developed over many years.

According to Harvard archives, AI was first formally discussed in the 1950s, not too long after Jean Rostand made his observation about our god-like capabilities. From early chess playing machines, it has since grown enormously. Automated chatbots synonymous with customer service support lines – which have frustrated consumers for decades – are a form of artificial intelligence. So too is Siri, Apple’s virtual assistant that was first launched back in 2011.

This is all to say that we’ve already been benefiting from AI for a very long time. And while criminals have used AI-backed tools during this period, we’ve also seen AI’s awesome ability to improve lives.

In healthcare alone, artificial intelligence has been able to beat radiologists when detecting various forms of cancer upon reviewing patient scans, it has been able to surpass a doctor’s ability to diagnose Parkinson’s disease based on breathing patterns, and it is being used to develop treatment plans for those who suffer from ultra-rare diseases that receive less attention due to their anomalous nature.

This aligns with the perspective of Carl Sagan, a great champion of science, when he observed, “advances in medicine and agriculture have saved vastly more lives than have been lost in all the wars in history.” Let us not then pause in our development of artificial intelligence.

Science may have made us gods before we were worthy of being human. But we stand to benefit far more from AI than we stand to lose. Granted this is an imperfect proposition, but it is therefore a human proposition.

--

Co-authored by David Brendel and Ryan Stelzer, Think Talk Create: Building Workplaces Fit for Humans was published by the Hachette Book Group under the PublicAffairs imprint on September 21, 2021. Now available to order!